Effects Matrix and Explanatory Effects Matrix

Description

Matrices that set out the outcomes of a programme or change, split by type and time, and then trace them back to who did what and why.

An effects matrix focuses on dependent variables: the outcomes of something. Because qualitative outcomes are hard to pin down, the analyst first decides what makes up the outcome. In the book's example 'organisational change' was split into structural, procedural and relational or climate changes (rows), each shown for early and later use (columns) and divided into primary changes, which follow directly from the innovation's requirements, and spin-offs, which are secondary effects not fully anticipated. Reading it shows how effects ripple out over time. The explanatory effects matrix asks why. Rows are sources (for example superintendent, building administrator, developers, peer users, materials); columns show the user's assessment of each source, what was provided, short-run effects in users' words, longer-run consequences, and a researcher explanation. It is a first step towards causal explanation, but each row is read as a separate sequence, so interactions between rows need case dynamics matrices or causal networks. For UX, these matrices organise the consequences of a launch, migration or policy change, including unintended side-effects, and connect them to what different support channels actually did.

Attributes

Loss type
Latent
Effort
Days

Steps

  1. Define the outcome of interest and break it into types (for example workflow, team structure, attitudes).
  2. Choose time columns (early versus later use) and, if useful, split each into primary changes and spin-offs.
  3. Enter outcomes from coded notes as short phrases.
  4. For explanation, build a second matrix with sources of influence as rows and columns for what was provided, short-run effects, long-run consequences and an analyst explanation.
  5. Fill cells with quotes and paraphrases, then compare rows to see which sources produced which effects.
  6. Note interactions the matrix cannot show and move to a case dynamics matrix or causal network if needed.

Sources

  • Miles, M. & Huberman, A. M., Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed. (1994)

Also known as

  • Effects matrix
  • Explanatory effects matrix
  • Outcomes matrix

Categories

  • User Research
  • Qualitative Analysis & Synthesis